Mind the Pool: Convolutional Neural Networks Can Overfit Input Size

Bilal Alsallakh, David Yan, Narine Kokhlikyan, Vivek Miglani, Orion Reblitz-Richardson, Pamela Bhattacharya

Keywords: Convolutional Neural Networks, Pooling, Input Size, Overfitting

TL;DR: Standard pooling arithmetic can cause CNNs to overfit the input size used during; an adjustment improves generalization to arbitrary sizes and robustness to translation shifts.

Abstract: We demonstrate how convolutional neural networks can overfit the input size: The accuracy drops significantly when using certain sizes, compared with favorable ones. This issue is inherent to pooling arithmetic, with standard downsampling layers playing a major role in favoring certain input sizes and skewing the weights accordingly. We present a solution to this problem by depriving these layers from the arithmetic cues they use to overfit the input size. Through various examples, we show how our proposed spatially-balanced pooling improves the generalization of the network to arbitrary input sizes and its robustness to translational shifts.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning